ISPA Account

2018 ICPA Photos

Suggestions For 2020 Conference Location
Have an idea for the location of the 2020 conference? Let us know by filling out this form.
Select the city to host the 15th ICPA Conference.
St. Louis, Missouri, USA (In conjunction with InfoAg 2020)
Minneapolis, Minnesota, USA
If OTHER selected, please enter the name of the city.
Enter the name and email of the contact for this city.
Enter your reason for suggesting this city for the 2020 conference.
Thank you for your time and interest. Your responses have been transmitted succesfully.


Access to the proceedings of the 14th ICPA is available online:
All the oral and poster presentation abstracts are available to the public. Access to the full papers and extended abstracts is limited to ISPA members and registered attendees of the 14th ICPA. Full papers and extended abstracts are also available for a fee to the public.
ISPA members need to login to the website using the panel on the left side of the home page. Use your email address associated with your membership.
If you attended the 14th ICPA Conference and do not know your password, or have not yet set up an account, you can click here.

Proceeding Spotlight

14th International Conference on Precision Agriculture Presentation

Real-Time Fruit Detection Using Deep Neural Networks

Proximal imaging using tractor-mounted cameras is a simple and cost-effective method to acquire large quantities of data in orchards and vineyards. It can be used for the monitoring of vegetation and for the management of field operations such as the guidance of smart spraying systems for instance. One of the most prolific research subjects in arboriculture is fruit detection during the growing season. Estimations of fruit-load can be used for early yield assessments and for the monitoring of harvest and thinning. In addition, the visual aspects of fruits enable to appraise their growth and ripening status. This paper proposes a new approach for real-time fruit detection, combining a fast geometrical pre-processing whose output feeds a deep neural network (DNN) classifier. The first step is a radial Hough-like operator, which aims at identifying quickly the regions of interest, restricting the use of the DNNs to the most probably genuine candidates. The proposed method is generic enough to be applied on most near-spherical fruits. It was tested in two contexts: grapes and apples, with different varieties and phenological stages. In both cases the proposed method provided promising results. Correlation coefficients with manual counting and real harvest loads are up to 0.96 for grapes and up to 0.85 for apples.